in privacy_lint/attack_results.py [0:0]
def _get_balanced_scores( scores_train: torch.Tensor, scores_test: torch.Tensor